according to a government researcher at Japan Atomic Energy Agency (JAEA).
Yomiuri Shinbun (9/22/2011):
The simulation done by Yasuteru Shibamoto, researcher at Japan Atomic Energy Agency, shows that the Containment Vessel of Reactor 2 at Fukushima I Nuclear Power Plant may have been damaged, and had a hole about 7.6 centimeters in diameter right after the March 11 earthquake.
It is the first time that the degree of damage on the Containment Vessel is estimated in numbers. It was announced on September 21 at the fall conference of the Atomic Energy Society of Japan in Kitakyushu City.
In simulation, Shibamoto utilized the data TEPCO had announced regarding the operation of the reactor core isolation cooling (RCIC) system and the change in pressure [inside the Containment Vessel]. In 14 hours after the earthquake hit, the water supply to the RCIC was switched from the condenser storage tank that had run dry to the suppression pool at the base of the Containment Vessel.
As the heat wouldn't dissipate, the pressure should have risen to about twice the design pressure (the design pressure is about 5 atmospheric pressures) within 2 days after the quake. However, in the actual data, the rise of the pressure was gradual, and it took more than 3 days to reach 7 atmospheric pressures.
I see. So when the water source was switched from the tank to the suppression pool, the pressure didn't rise because the Containment Vessel already had a hole in it. Then the suppression pool blew up on March 15.
The article doesn't say where the hole might be, or exactly when it may have been created and how.